Elsevier

Health Policy

Volume 106, Issue 3, August 2012, Pages 246-256
Health Policy

General practitioners’ preferences for the organisation of primary care: A discrete choice experiment

https://doi.org/10.1016/j.healthpol.2012.03.006Get rights and content

Abstract

Objectives

To examine GPs’ preferences for organisational characteristics in general practice with focus on aspects that can potentially mitigate problems with GP shortages.

Methods

A simple random sample of 1823 GPs (corresponding to half of all GPs in Denmark) was drawn at the beginning of 2010, and a response rate of 68% was obtained. A discrete choice experiment (DCE) is applied, and attributes included are: practice type (solo/shared), number of GPs in general practice, collaboration with other practices (yes/no), change in weekly working hours (administrative versus patient related) and change in yearly surplus. Multinomial logit analyses (with and without interaction variables) are used, and marginal rates of substitution are calculated.

Results

GPs working in solo practices have different preferences for the organisational attributes compared to GPs in shared practices. The compensation needed for GPs to re-organise from solo to shared practice is associated with the size of the practice. GP characteristics such as age, working hours and surplus affect their willingness to undergo organisational changes.

Conclusions

Our results are of relevance to decision makers in designing policies aimed at influencing GPs’ organisation in order to overcome problems related to shortages.

Introduction

In recent years, most European countries have experienced a lack of general practitioners (GPs) due to an ageing GP workforce and an increase in demand for more flexible positions and for part-time work [1]. At the same time, the demand for health care services is increasing due to ageing populations, greater occurrences of chronic diseases and developments of new methods of treatment [2]. Various proposals have been put forward to address how to avoid the mismatch between the demand and supply of health care services in general practice, and different strategies can be pursued to bring down demand and/or to increase supply. This study focuses on the supply side, where resources can be supplied, e.g. by recruiting new GPs and/or trying to maintain the already established GPs in general practice. For GPs already settled in general practice, resources can be released by sharing practice personnel, shifting tasks from GPs to nurses and other health care professionals [3], [4], or by inducing GPs to work longer hours. To enable GPs to shift tasks to other parties it is recommended by health authorities that GPs re-organise in larger shared practices, in order to mitigate some of the problems related to GP shortages and to increase efficiency [5].

In Denmark the focus has in recent years been on addressing issues related to GP shortages and quality improvement in primary health care [6]. The issue of re-organisation is especially pertinent in Denmark, where GPs run private practices, either as solo or shared practices, and where as many as one third of all GPs work in solo practices. More than 70% of GPs work in practices with one to three GPs per practice while only very few practices in Denmark have seven GPs or more. Solo practices as well as shared practices can organise in a collaboration practice, which involves collaboration and cost sharing (sharing of facilities, equipment and/or personnel), but excludes sharing of revenues. Approximately one fifth of all practices in Denmark are organised in a collaboration practice [7]. In 2010 the Organisation of General Practitioners in Denmark estimated a lack of GPs of approximately 100 (corresponding to 3% of the GP population). Moreover, it was anticipated that the GP shortage would expand in the coming 3–4 years [8].

In this study a discrete choice experiment (DCE) is conducted to investigate the GPs’ preferences for organisational characteristics of general practice in order to gain knowledge of which aspects are deemed most important by GPs in the organisation of primary care. The focus is on aspects that can potentially mitigate problems with GP shortages, which is valuable knowledge in countries with a scarcity of GPs. More specifically, GPs’ preferences for type of practice, number of GPs in practice, working hours and content of working hours are explored. The study provides knowledge of which aspects are important to the GPs and how much (if at all) GPs should be compensated to release resources and re-organise.

The knowledge of GPs’ preferences for organisational aspects is limited, and only a few studies have investigated this topic by means of stated preference methodologies. In a DCE study, GPs’ preferences for pecuniary and non-pecuniary job characteristics in the UK are elicited [9]. The study found that particularly hours of work commitment was an important attribute for the doctors as were the pecuniary factors. It was shown that preferences differed between sub-groups of GPs, which led to the conclusion that it would require different policies to influence the choice of practice significantly. Likewise, preferences for general practice jobs in Scotland for sessional and principal GPs were explored using the DCE [1]. It was observed that sessional GPs attached lower values to consultation length and hours of work, and that professional development was less important than for principal GPs. The study concluded that one general contract would fail to cater for all GPs. In another DCE study, preferences of recently appointed GPs in South-east England were investigated. It was found that the most important influence on GPs’ choice of practice was aversion to location in an area of high deprivation and that GPs should receive a compensation of just over 5000 GBP per year to be willing to work in an area with a high proportion of deprived patients [10]. In an on-going longitudinal survey of the GPs’ working lives in Australia, the DCE is used to elicit GPs’ preferences for different job characteristics. The characteristics included in the experiment are earnings, working hours, on-call arrangements, location, social interaction, possibilities of a locum and average consultation length [11]. In addition, a number of job satisfaction surveys have been conducted for GPs in the UK including some organisational characteristics such as working hours, physical working conditions, remuneration, etc. [12], [13], [14]. These studies found among other things that extended possibilities of flexible working hours would be valuable; that job and personal characteristics of the GPs have an effect on job satisfaction; and that the most frequent factors of job stress was increasing workload, paperwork and insufficient time to do justice to the job. The Organisation of General Practitioners in Denmark conducted a GP satisfaction survey in 2008 in order to investigate GPs’ attitudes towards organisational issues such as opening hours, administrative work, reimbursement, etc. Results from the survey showed that GPs were willing to increase the number of patients on their list if they were duly compensated, but that they were not willing to increase opening hours. The preferences for larger shared practices were not investigated directly, but generally there was no interest in increasing collaboration. This was true for GPs in solo and shared practices. The GPs generally preferred fewer administrative tasks and more time for patient consultations [7].

DCE studies (as opposed to satisfaction studies) enable researchers to systematically investigate the relative importance of the included topics of investigation since the methodology forces the respondents to make trade-offs. DCEs hereby provide much more information about the importance of different aspects relative to each other. This study adds to the literature on GPs’ preferences for the organisation of primary care and takes on a novel focus where the preferences of GPs are tested specifically with respect to universal organisational characteristics which can help mitigate problems related to GP shortages. The study is the first to explicitly exploit preferences for organisational characteristics that are directly related to overcoming problems with GP shortages. The job characteristics included in the preference task are deemed universal and results may be generalised to other countries with similar problems of GP shortages.

Section snippets

Survey design

The issues and questions to be considered in the survey were carefully chosen on the basis of (1) earlier studies from other countries with similar problems in the primary care sector, (2) three whole day observational visits at general practices (one. solo, one shared and one collaboration practice), (3) interviews and discussions with GPs and (4) discussions with The Organisation of General Practitioners in Denmark and Danish Regions. Policy issues dealing with the potential problem of GP

Results

Of the 1823 distributed questionnaires 1235 were returned, which results in a response rate of 68%. The collected sample is representative of the Danish GP population with respect to age, gender and practice type, except that GPs in the age group 60–69 are slightly underrepresented in the sample [25]. Table 2 shows the descriptive statistics for selected GP and general practice characteristics as well as the responses to some of the attitudinal questions in the questionnaire. With respect to

Discussion

The focus of the present study was to elicit Danish GPs’ preferences for a re-organisation of general practice, a topic that is relevant given that GP resources are scarce and the demand for GP services is increasing. Preferences for a number of characteristics that can release resources have been investigated with a particular focus on willingness to increase working hours and solo practitioners’ willingness to re-organise in larger practices. Currently, there is political focus on promoting

Concluding remarks

This study has shown that GPs working in solo practices have different preferences for organisational attributes compared to GPs in shared practices. In particular, we find that the compensation needed for GPs to re-organise from solo to shared practice is strongly associated with the size of the practice. Further, we show that the GP characteristics such as age, working hours and surplus affect their willingness to undergo organisational changes. Overall, our results provide important

Acknowledgements

Thanks to Thore Eriksen, Jesper Lykkegaard and Adam Moltke for their useful comments and suggestions.

References (26)

  • A. Richards et al.

    Skill mix between nurses and doctors working in primary care-delegation or allocation: a review of the literature

    International Journal of Nursing Studies

    (2000)
  • A. Scott

    Eliciting GPs’ preferences for pecuniary and non-pecuniary job characteristics

    Journal of Health Economics

    (2001)
  • S. Wordsworth et al.

    Preferences for general practice jobs: a survey of principals and sessional GPs

    British Journal of General Practitioners

    (2004)
  • M. Strandberg-Larsen et al.

    Denmark: health system review

    European Observatory on Health Systems and Policies

    (2007)
  • M. Laurant et al.

    Substitution of doctors by nurses in primary care

    Cochrane Database of Systematic Reviews

    (2005)
  • H. Charles-Jones et al.

    Transforming general practice: the redistribution of medical work in primary care

    Sociology of Health and Illness

    (2003)
  • Danske Regioner

    Organisering af almen praksis. Underrapport: Almen praksis i fremtidens sundhedsvæsen

    (2007)
  • Praktiserende Lægers Organisation

    Medlemsundersøgelse marts 2008

    (2008)
  • Praktiserende Lægers Organisation

    Faktaark om almen praksis

    (2010)
  • T. Gosden et al.

    How do general practitioners choose their practice? Preferences for practice and job characteristics

    Journal of Health Services Research and Policy

    (2000)
  • Medicine in Australia: Balancing Employment and Life (MABEL)

    General Practitioner & GP Registrar

    (2008)
  • F. French et al.

    General practitioner non-principals benefit from flexible working

    Journal of Health Organization and Management

    (2005)
  • A. Scott et al.

    Job satisfaction and quitting intentions: a structural model of British general practitioners

    British Journal of Industrial Relations

    (2006)
  • Cited by (17)

    • Does accreditation affect the job satisfaction of general practitioners? A combined panel data survey and cluster randomised field experiment

      2020, Health Policy
      Citation Excerpt :

      GPs are contracted with the health care authorities (Danish Regions) under a mixed capitation and fee-for-service system [22]. Approximately two thirds of general practices are organised as single-handed practices employing one third of GPs [23]. In 2016, there were 3421 GPs in Denmark working in 1922 practices ([24]; [25]).

    • The impact of pecuniary and non-pecuniary incentives for attracting young doctors to rural general practice

      2015, Social Science and Medicine
      Citation Excerpt :

      which initially allowed respondents to reflect freely on the question. The topics were chosen in alignment with the study and policy objectives and inspired by earlier studies (Harris et al., 2005; Pedersen and Gyrd-Hansen, 2013; Pedersen et al., 2012; Scott, 2001), and the ongoing MABEL study (Sivey et al., 2012; Scott et al., 2013). The qualitative work resulted in the selection of the following two sets of attributes to be included in the DCE: (1) work-life attributes (i.e. location, opportunity to control working hours, and income) which affect the doctor's out-of-work life, and (2) intrinsic attributes which relate to characteristics of experiences of the workplace (i.e. practice size and opportunity for professional development) (Sivey et al., 2012).

    View all citing articles on Scopus
    View full text